stones on beach during sunset

Belonging (Part 2: The tide may be turning)

I am tired of being in a position where I have to apologize for sexism, racism, and other toxic ideologies within this movement. I am tired of convening calls with other community builders where we try to figure out how to best react to the latest Thing That Happened. I am tired of billionaires. And I am really, really tired of seeing people publicly defend bad behavior as good epistemics.

Megan Nelson, “I am tired
Listen to this post

1. Introduction

On January 12th, Anders Sandberg (Senior Research Fellow, Future of Humanity Institute) posted an apology to Twitter on behalf of Nick Bostrom for remarks that Bostrom made to a listserv in the mid-90s.

Although the existence of an apology is a step in the right direction, we have seen that both the original email and the later apology were problematic for a number of reasons.

Today, I want to talk about the reaction to Bostrom’s email within the effective altruism community. I think that from January 12th to January 14th, much of that reaction was concerning. Today (January 15th) I am beginning to see signs of hope, as the tide turns against the worst discussion and defense of racist pseudoscience and towards repudiation, open admission of guilt, and a genuine commitment to learning and change.

Below, I describe five bad reactions to the Bostrom scandal and five better reactions that are beginning to emerge and counteract these initial bad reactions. I describe how the tide is showing signs of turning in order to highlight the possibility for positive change within and outside of the effective altruism movement. I hope that the movement can continue to take steps in the right direction, so that some good may come out of an otherwise regrettable incident.

As before, I apologize for the rushed nature of this post. We academics are not trained to write quickly, but sometimes speed is of the essence. I hope that my thoughts will prove useful and my errors minimal.

2. Bad reactions

On January 12th, 13th and much of the 14th, a significant segment of the effective altruist community reacted badly to the Bostrom scandal. In this section, I will describe five bad reactions that were prominent at this time, give evidence for their prominence, and explain why these reactions are wrong.

The first bad reaction is open discussion, or even endorsement of discredited racist pseudoscience. On January 13th, a comment was submitted to my blog directly citing the racist writings of Dr. Philippe Rushton and Dr. Arthur Jensen to argue that we should leave open the possibility of a significant IQ gap between racial groups grounded in underlying genetic causes. (This work has been widely repudiated by leading scientists, and by Rushton’s own department).

In the background looms a poll by the popular blog Slate Star Codex, which asked its readers in 2020: “How would you describe your opinion of the idea of “human biodiversity”, eg the belief that races differ genetically in socially relevant ways?” (1 = very unfavorable, 5 = very favorable). The existence of this poll showed a callous willingness to entertain racist pseudoscience as a topic for discussion, and the results of this poll showed a disturbing level of support for racist views:

The same question appears on the 2022 ACX survey, which is still open. It appears that this corner of the community still regards the theory of “human biodiversity” as a legitimate topic for discussion.

Defenders of “human biodiversity” theory made sizable inroads on major discussion fora. Here is the EA Forum on January 13th (note the agreement votes):

And here is LessWrong today (this comment has positive, though thankfully falling Karma and agreement):

Why this is wrong: Racist pseudoscience has been the driving force behind many of the worst atrocities in recent history, and continues to be a major motivation for discriminatory and harmful policies. Operating under a false cloak of objectivity, a commitment to pseudoscientific beliefs contravenes the community’s explicit commitment to using reason and evidence to set global priorities, illustrating a striking willingness to disregard reason, evidence and the unanimous voice of the scientific community in a motivated attempt to sustain racist beliefs.

It is wrong not only to hold racist beliefs, but also to treat racist pseudoscience as a serious topic for discussion. One reason why this is wrong is that it falsely implies that the relevant science has not already been refuted and helps to sustain scientific discussions that should have ended long ago. Another reason why this is wrong is because discussions of racist pseudoscience play an ongoing role in sustaining oppression. In fact, we will see that racist pseudoscience played a role in almost all of the other bad reactions described below.

A second bad reaction is defending the indefensible. There are many reasons why both Bostrom’s original email and his subsequent apology were wrong. Despite this, Bostrom attracted a chorus of defenders unwilling to concede that Bostrom had done anything terribly wrong.

Consider this reaction on January 12 to CEA’s condemnation of Bostrom:

Or this reaction, which repeats Bostrom’s slur (which I have redacted) and has still not been removed by the EA Forum.

Why this is wrong: Beyond the stunningly inappropriate utterance of a racial slur, defending Bostrom is wrong for a number of reasons. First, Bostrom’s behavior was wrong, and it is appropriate to condemn misbehavior. Second, a defensive stance stands in the way of learning, growth and change. We cannot have a productive and forward-looking discussion about how to improve beliefs and behavior within the effective altruism community until the community comes to publicly understand and accept the existence of a problem.

A third bad reaction is failure to acknowledge the significance of the problem. On January 14th, Bostrom edited his website to claim that the condemnation of his behavior is a distraction from what is really important.

Why this is wrong: Anti-racism is important. Racist pseudoscience has caused, and continues to cause a tremendous amount of direct and measurable harm to some of the most vulnerable communities in the world today. Furthermore, prejudiced views shape how we envision the future and threaten to worsen the future that we will work to create. We need to see racism as an important problem before we can take active steps to lessen its prevalence and impact in society.

A fourth bad reaction is silence. Far too many effective altruists have simply said nothing about the Bostrom affair. Although we will see below that the voices of condemnation are growing, there are still far too few of them.

Why this is wrong: Many effective altruists are young, vulnerable, and in a position where it may be seriously unsafe for them to speak out. It is not wrong for them to stay silent. Those of us better able to absorb the costs of speaking out will speak in the meantime.

But those in a position to speak out have a moral duty to use their position to drive positive change in the world. If effective altruists are serious about using reason and evidence to do good better, they must raise their voices to protest the misuse of reason and the utter lack of convincing evidence being marshaled in favor of discredited and racist views.

The Archbishop Desmond Tutu held that “If you are neutral in situations of injustice you have chosen the side of the oppressor.” Injustice thrives on the silence of those who would stop it and shrinks from the light of truth, reason and earnest scrutiny. Speaking out will help to build a better and more just future.

A fifth bad reaction is a refusal to learn, change and grow. We saw in our original discussion that Bostrom suggests the incident should be buried:

The best thing would probably have been for all of those words to have been buried and forgotten. Good ideas have been taken up in other contexts, further developed, and written up in more readable formats; and the no-good stuff deserves to rest in the great ash heap of history.

Bostrom defends this in part by suggesting that his views were never problematic in the first place: his apology held that the email “does not accurately represent my views, then or now”.

Why this is wrong: Many of us are privileged to notice racism only when it becomes so visible that it is impossible to ignore. This means that it is crucial to exploit the fleeting moments in which problematic attitudes become visible to all as opportunities to learn about the harmful beliefs and behaviors which have been so long ignored; to change those beliefs and behaviors; and to grow to be more aware of, and responsible for our behaviors in the world and their effects on those upon whom the consequences of racist thought and talk are heaped every day.

3. Good reactions

Recently, there have been signs that all of the above reactions are beginning to be repudiated and reversed. This gives me hope.

First, instead of open discussion and endorsement we now see open repudiation of the contents of racist pseudoscience as well as a refusal to treat racist pseudoscience as a suitable topic for discussion. The author of the comment submitted to my blog emailed me on January 14th to apologize for citing Rushton and Jensen and to repudiate any claims about a genetic basis for differences in intelligence. This is exactly the type of growth- and learning-minded response that can bring lasting, positive change within the movement.

Although Slate Star Codex shows no signs of change, Mohammad Ismam Huda’s comment on the EA Forum now receives sizable support, and is accompanied by a well-received debunking of racist pseudoscience. (I personally would not have posted that debunking, because I think it goes too far towards treating racist science as a legitimate topic for investigation, but I appreciate the community’s reaction to this post).

Although there has been strong progress in the community’s willingness to denounce racist pseudoscience as false, there is continued room for progress in the community’s understanding of the need to reject discussion of harmful and repudiated scientific studies. Sometimes, that rejection is clear. This is current on the EA Forum as of the afternoon of January 15th:

That is good. And Richard Chappell gestures in the same direction:

But this is also current on the EA Forum:

The same post appears on LessWrong, with substantially more upvotes. This needs to be removed.

A second good reaction is publicly condemning the indefensible. CEA acted well in this regard, issuing the following statement on January 12th:

On the same day, Habiba Islam (now of 80,000 Hours and formerly my colleague at the Global Priorities Institute) issued a very public personal response to Bostrom’s words:

Today, my colleague Hayden Wilkinson (Postdoctoral Research Fellow, GPI) said the following:

These statements (along with many others) showed courage, honesty and a commitment to learning and change.

A third good reaction is acknowledging the significance of the problem. A post by EA community-builder Megan Nelson, released quickly on January 12th said:

Earlier today, Nathan Young made an earnest attempt on the EA Forum to voice what he calls the “this is very bad” position (albeit not in his own voice):

And another comment, again on the EA Forum, hit the nail on the head:

I would like to see more acknowledgment of the seriousness of the problem in the near future, and I would be especially heartened to see these thoughts being uttered in full voice, rather than as an attempt to reconstruct the reactions of others. But all of these posts show a marked improvement over previous refusals to acknowledge the seriousness of the problem.

A fourth good reaction is speaking out. Today on the EA Forum, Rohit Krishnan urges us to “Do better, please.”

Yesterday, Robert M called on the EA Forum to “speak the truth, even if your voice trembles.”

These fears are real. Last November, Linda Kinstler of The Economist had this to say:

Over the past two years, I’ve heard many stories of young, ambitious people who came to effective altruism wanting to change the world but grew disenchanted. Many people I spoke to didn’t want to be identified, concerned that the community might retaliate by reducing funding or offering fewer professional opportunities.

I must confess that my voice is trembling. Never in my life have I so openly bitten the hand that feeds me, not even when (along with the graduate students’ union at Harvard) I took to the streets with picket signs against my own employer.

I am young, actively on the job market, in need of money, employment, friendship and support and deeply concerned that I may lose many or all of these things by speaking out.

Nevertheless, I firmly believe that if you are in a position where it is safe to speak out, then speaking out is not only necessary and beneficial, but in fact morally obligatory. I would urge more people to speak their minds if they are in a position to do so.

A fifth good reaction is a commitment to learning, change and growth. Some of the good reactions mentioned above have yet to make this commitment. For example, my largest objection to what is otherwise one of the better statements made by EAs so far (the CEA statement) is that it does not promise any change.

However, we saw that many others have called for change. Habiba Islam and others call for increased sensitivity in the way that race is discussed within the community. Hayden Wilkinson calls for change in the position of Nick Bostrom within the community. My own organization is coordinating with other organizations based in Trajan House to prepare an institutional response of a kind likely to lead to lasting change.

As we move forward into the coming week, I hope to see more effective altruists acknowledging the need for broader change around the way that considerations related to race, gender and other socially-constructed categories are wielded within discussions, with consequences for the standing of vulnerable members of the community and for the world.

4. How the conversation must proceed

In introducing her book Why I’m no longer talking to white people about race, the journalist and author Reni Eddo-Lodge explains what usually happens when she tries to raise questions of race in discussion:

I can no longer engage with the gulf of an emotional disconnect that white people display when a person of colour articulates our experiences. You can see their eyes shut down and harden. It’s like treacle is poured into their ears, blocking up their ear canals like they can no longer hear us … Their eyes glaze over in boredom or widen in indignation. Their mouths start twitching as they get defensive. Their throats open up as they try to interrupt, itching to talk over you but not really listen, because they need to let you know that you’ve got it wrong. Even if they can hear you, they’re not really listening. It’s like something happens to the words as they leave my mouth and reach their ears. The words hit a barrier of denial and they don’t get any further.

This suggests the importance of listening as a prerequisite to successful conversation, even (indeed, especially!) when listening is hard.

I can no longer have this conversation, because we’re often coming at it from completely different planes. I can’t have a conversation with them about the details of a problem if they don’t even recognise that the problem exists.

In addition to listening, it is important to guard against the inclination to dismiss and retaliate against those who raise uncomfortable questions. Eddo-Lodge continues:

Entering into conversation with defiant white people is a frankly dangerous task for me. As the hackles rise and the defiance grows, I have to tread incredibly carefully, because if I express frustration, anger, or exasperation at their refusal to understand, they will tap into their pre-subscribed racist tropes about angry black people who are a threat to them and their safety. It’s very likely that they’ll then paint me as a bully or an abuser. It’s also likely that their white friends will rally round them, rewrite history and make the lies the truth. Trying to engage with them and navigate their racism is not worth that.

This situation, Eddo-Lodge says, presents her with two terrible options: “speak your truth and face the reprisal, or bite your tongue and get ahead in life.” After many years of fighting, Eddo-Lodge has tired of the reprisals. I hope that effective altruists will not drive other much-needed voices out of public discussion through defensiveness, emotive criticism or retaliation.

If we can learn to listen (really, genuinely listen) without dismissal, anger, or threat of reprisal, we can, all of us, take this as a teaching moment to learn about the continuing role of race within society today and to use that newfound knowledge to drive positive change.

Comments are open, but please tread carefully. I would urge readers to keep in mind the above remarks about how the conversation must proceed as well as the blog’s comment policy. Those we speak to and speak about are real people. Please treat those you speak to and speak about like real people who matter just as much as any other person on this planet.

Under no circumstances will I publish any sort of speculation about the relationship between race and intelligence.


Posted

in

by

Comments

6 responses to “Belonging (Part 2: The tide may be turning)”

  1. Esben Kran Avatar
    Esben Kran

    Thank you very much for voicing these much-needed concerns and your efforts towards making long-lasting change! I and many I talk with are quite disenchanted with many recent aspects of effective altruism (even as EA-funded organizations) and Bostrom has not alleviated this in any capacity.

    I also urge you to read the article against cancelling, though I am quite open to the idea of Bostrom stepping down: https://forum.effectivealtruism.org/posts/zYNDJxDm4tWgw8AWn/avoiding-munich-s-mistakes-advice-for-cea-and-local-groups

    More generally, borderline illegal behaviour in my opinion should lead to “cancellation” of that individual’s participation in specific types of events. I’ve had several talks with women about situations where e.g. sexual harassment is similarly treated in a very bad manner within some Bay Area-related EA circles.

    1. David Thorstad Avatar

      Hi Esben!

      Thank you for your much-needed voice of disenchantment with the current situation as well as for raising the important issue of how Bostrom should be treated. (And thank you for a splendid example of the style in which I hope our conversations about this issue can be conducted).

      I have not recommended that Bostrom should resign, or for that matter made any definite recommendation for how Bostrom should be treated. I have not done so for three reasons.

      First, I do not think it is my place to legislate how Bostrom should be treated. That is a decision for the community, and frankly for the world.

      Second, I think that many people’s reactions may depend on what Bostrom does next. There is always room for healing, learning and growth. If Bostrom provides strong evidence of a genuine commitment to lasting growth and change, a repudiation of racist science, and an apology for harms caused by past words and deeds, that may go some way towards softening many people’s reactions. (Conversely, it is also possible to harden our hearts). But again, I have no right to speak for everyone.

      Finally (and most importantly), I think that to focus on the question of how Bostrom should be treated is to ignore the larger issue. There is clear evidence of a systemic problem that needs to be fixed, and I hope that we can harness the energy generated by this incident to begin much-needed conversations and initiatives that will lead to lasting change. Others may be more focused on punishment, and that is their right. My own hope is that effective altruists may take this as a moment to learn how to do good better.

  2. Bob Jacobs Avatar

    > and is accompanied by a well-received debunking of racist pseudoscience. (I personally would not have posted that debunking, because I think it goes too far towards treating racist science as a legitimate topic for investigation, but I appreciate the community’s reaction to this post)
    Yeah, I blundered pretty hard writing that debunking, but not just because of its legitimizing effect… When you write such a thing on a public forum, you put yourself on the hook to keep responding to any comment that claims to debunk what you say. I tried to get ahead of this by requesting that if anyone thought they spotted a flaw in my list, they send me a direct message, but obviously, this got ignored. So then the question becomes: how much time do you want to invest to ensure that the last comment on the thread isn’t one claiming the anti-racism arguments just got debunked?

    Ways to get around this:
    1) Direct them to a different forum. Instead of debunking it yourself, link to someone else’s debunking that has a comment section so they can comment there instead. Preferably, link to someone who enjoys and gets energized by debunking commenters instead of getting dejected and drained from it. This might work for some commenters, but others won’t be redirected, so maybe also…
    2) Precommit to not respond to any replies. This saves you time but doesn’t look too good to an outsider, as it might read as someone who knows they’re wrong cowardly avoiding the conversation. So maybe instead…
    3) Precommit to a limited number of replies. Something like: I will make time to answer ten comments and no more. That way, it doesn’t look like you’re avoiding criticism and makes your exit from the conversation reasonable.

    These tips work for any forum, but I also have a further tip for doing this on the EA forum specifically. The EA forum has a voting-power-by-popularity system. If you get upvotes, you get more voting power; if you get downvotes, you get less voting power (yes, this does contribute to the groupthink in EA). Even worse, a comment that’s downvoted doesn’t appear on the front page where other users can normally see it. Also, if someone leaves a reply to your comment, you get a notification. You can see where this is going. When I replied to the HBD proponents, they got a notification and would immediately downvote me, deleting it from the front page. Then they would mass downvote me, including older comments and posts, ensuring I lost voting power.

    So the obvious solution is to use an anonymous account to post any criticisms that EAs might downvote. That way your main account doesn’t lose voting power. If you were wondering why so many EAs feel the need to leave their criticisms anonymously, this is part of the reason why.

    1. Bob Jacobs Avatar

      > So the obvious solution is to use an anonymous account to post any criticisms that EAs might downvote.

      I wanted to clarify that this is the solution for the individual, not for the movement. The fact that people have to use anonymous accounts is bad for the movement in a couple ways.
      Firstly, it fosters a culture of fear and mistrust. When people feel they must anonymize their dissent, it signals that the environment is not open to healthy debate, which stifles intellectual growth and collaboration.

      Secondly, anonymity undermines the credibility of criticisms. It becomes difficult to discern whether a critique comes from someone within EA with relevant experience and goodwill or from an outsider lacking context or acting in bad faith. As a result, legitimate criticisms are more easily dismissed, and the movement risks fostering the appearance of a false consensus—where all the well-known EAs seem to support the trajectory of the movement uncritically.

      Thirdly, the reliance on anonymity can harm the movement’s external reputation. Critics outside EA may interpret the need for anonymity as evidence that the movement is intolerant of dissent or overly insular. This perception could alienate potential allies and make it harder to attract thoughtful, independent-minded individuals to join or support EA. Or worse, if EA keeps saying how great they are when it comes to criticisms, but then outsiders see that so many people feel the need to use anonymous accounts, they may think EAs are deluding themselves, or worse, lying.

      Fourthly, anonymity reduces accountability for both critics and those receiving criticism. Critics may feel less responsible for the content and tone of their arguments, leading to unproductive or overly harsh feedback. Conversely, EA leaders may feel less compelled to engage meaningfully with anonymous criticism, as the lack of identifiable authors makes it easier to dismiss or ignore.

      Lastly, if you create an anonymous account for your criticisms this account will have less voting power, meaning the praise on your main account stays longer on the front page than the criticism on your anonymous account. This skews the perception of consensus even further, as people are exposed to fewer dissenting voices.

    2. David Thorstad Avatar

      Thanks Bob!

      Apologies for the very late reply. I had flagged on my daily to-do list that I should reply to your first comment, then somehow never got around to it.

      I hope that those calling for myself and others to engage in serious object-level discussion of race science might take Bob’s experience to heart. Bob is a thoughtful person, no sworn enemy of effective altruism, and considerably more willing to engage in direct discussion of race science than I am. At the same time, Bob’s experience of trying to explain why racism is wrong was so terrible that Bob does not want to repeat it.

      The downvoting dynamic is real and important. It happens regularly to anyone attempting to challenge racism or sexism in the community. One of the easiest ways to suppress criticism is simply to ensure that it is not widely read or discussed, and the downvoting mechanism accomplishes this. This has the added benefit of being anonymous, and therefore not something that the individuals in question can be held accountable for.

      I think that you are exactly right, Bob, in worrying about a state in which people feel that they have to use anonymous accounts to post criticisms. I think that all of the downsides of this situation that you raise are important.

      I would also like to add another downside: when sharp criticisms are socially punished or made epistemically laborious, there is pressure for critics to moderate their views in order to incur less social sanction or take on less epistemic labor. In response to many of the most recent behavioral controversies within effective altruism, the most successful criticisms have generally been quite moderate – for example, pointing out that racism is disadvantageous instead of saying that racism is wrong. This makes it very tempting to advocate for relatively shallow changes, and the karma system is around to take care of those who do not.

      I don’t quite know what to do about this situation. When I think about behaviors that I do not like, such as open defense of racism, I must confess I find myself glad that people feel pressure not to engage in these behaviors or to do so under the cloak of anonymity, and I would not mind seeing these folks asked to do the epistemic labor of giving serious reasons for their views. I suspect that a great deal of readers’ perspectives on these issues may depend on where they sit with respect to the object-level issues being discussed.

      I think that effective altruists are very hesitant to legislate meta-level discourse norms based on particular object-level stances toward the views being discussed. If one person likes witchcraft and another thinks that witchcraft is evil, then many effective altruists are hesitant to impose discursive norms that rule out some potential defenses of witchcraft. I think it is important to think critically about the consequences of this stance. It is, as Scott Alexander might put it, a good way to end up with a coven of witches.

      1. Bob Jacobs Avatar

        > another downside
        Agreed
        > downvoting
        It’s mostly the fact that it affects your voting power that’s (imo) the problem. On the EA subreddit I feel quite comfortable providing a heterodox viewpoint, because even if they’re downvoted it won’t affect my next comment or post.
        I talk a lot about the effects of downvoting on EAF voting power, but the upvoting is also made a problem. You have the dynamic of people who praise popular EAs getting more voting power from them, which then leads to those praises being spread even further. Even if downvotes didn’t decrease voting power, this would still amplify groupthink. But also, if I see someone who talks about things I disagree with, and they say something I do agree with, I have to think about whether upvoting it will just spread their other messages I disagree with further (and obviously the same dynamic is also there with downvotes).
        This mechanism incentivizes making everything about the person, and whether they are “ingroup” enough, and not about the content of the actual post or comment.

Leave a Reply

Discover more from Reflective altruism

Subscribe now to keep reading and get access to the full archive.

Continue reading